Slashdot Mirror
New Display Technology to Compete with LCDs?
NetRanger writes "C|Net's News.com has a really interesting article to a new display technology that is based on interference of light patterns. The company, Iridigm, has a very compelling case for why their display method is far superior to LCD, including far brighter displays, far less power consumption... but the cool this is that the display actually works like RAM (it retains its state until voltage is applied to reset it) -- so what do you see when the driver crashes?"
- so what do you see when the driver crashes?"
Porn Screen Of Death?
That's gonna make shutting off the monitor real fast to hide the porn from your (wife/boss/Priest/Teacher) a lot more difficult.
Therefore, this tech will never fly.
So, if we can project this trend out, when do I need to start wearing sunglasses while I'm in the office?
You see? You see? Your stupid minds! Stupid! Stupid!
I've noticed that some frame buffers on laptops tend to retain images from other modes in memory till you go into that mode. So if I like crash my laptop looking at a pr0n site, reboot, when X starts, I will see what I saw till X redraws the screen... normally about half a second....
What speed? DDR? SDR? will it be adversely affected by magnetic fields? I know my LCD isn't phased by having my speaker right next to it, but my CRT sure as hell was... Will this thing be sensitive to EM?
---
Programming is like sex... Make one mistake and support it the rest of your life.
It seems like it would *look* beautiful, but would be costly to operate.
Of course, if you're going to shell out the cash for this, then you're probably not going to be worried about the electric bill.
Still sticking to my CRT for now...
If a and b in c, and a can create b, and a can create a, and b can create b, and b cannot create a, then a created c.
They're based on moving the membrane every time a pixel changes color. Wonder how many times you can do that before the membrane develops stress fractures.
Wonder if fractures would cause a failure, too.
I guess as long as it's at least as long as the expected useful life of an LCD backlight it's still a win.
The SCO lawsuit makes me wish my company were in Utah. We need a new building.
Hmm... No product displays at the website. Just some diagrams and a a photoshopped display.
That said, I'm currently tied to CRT technology because a lot of the media I have to deal with is color matched. Since color on a CRT screen is unreliable... it changes if you look at your screen from a different direction... this could offer a great deal of help to people like me who are tied to heavy, bulky displays rather than sweet flat-panels.
Of course the key here is that they have to deliver everything they promise in the way of omni-directional viewing and color-correctness.
The next Slashdot story will be ready soon, but subscribers can beat the rush and slashdot the links early!
If my computer crashes, leaving something unpleasant on my screen, can I clear it by picking it up and shaking it?
Not so much a sig as a lack of one.
Only like SRAM, not DRAM.
SRAM is pretty much static until changes are made, DRAM you'll hear described like a leaky capacitor. When you give it a charge it will slowly loose it, so you need to refresh it... many many times per second.
Help Brendan pay off his student loans
What I want is e-paper, paper that can be written on digitally, by people with the proper pens and machines with the proper hardware. This sounds a lot like a step in that direction.
My idea for the use of this paper is for notebook computers to be like scrolls. Initially just a tube, you pull out the screen which is rolled up inside (and has a rigid piece across the top), and unfold two braces (on both sides) to hold it in place.
They already have keyboards that you can roll up, why not screens? The scroll-book would do the same thing to store the keyboard as with the screen.
Persistence of images when the power goes off is a big requirement for digital paper. But I'm waiting for the scroll-book, which please note could double as a book and notebook if you could write on it with a digital pen. Don't unfurl the keyboard if you don't want to type into it.
Here's another second of thought...does anyone really think they'd announce a display technology that limits the user to a few femtometers of movement? Good lord...
The real potential comes when they can isolate sections of the screen to update. Since most screens remain, I would say, 80% the same, this could greatly increase the battery life of laptops since the screen is one of the largest power consumers. Isolating sections would allow only a small section to draw power when changed. The key would to make the sections as small as possible (pixel?) so that mouse movements don't cause un update to 1/4 the screen.
RTFA Genius
The iMoD elements are built upon two conductive layers--one a flexible metal membrane, the other is a thin film. These layers are held about 1 micron apart between two sheets of glass. When a voltage is applied to the element, the metal membrane layer becomes attracted to the thin film layer, turning the element black. Varying the voltage brings the layers closer and farther apart, and the distance between the layers determines what color--red, green or blue--the element displays.
Thus the only distance you have to control is between the membrane and film. Then unless you were moving at significant fraction of the speed of light the colors wouldn't change on your motion much.
Why, o why must the sky fall when I've learned to fly?
RTFA !
The display uses two plates on each pixel that can get closer or farther one from the other. The interference occur in the reflective part of the monitor, only to create the right frequency. Just like a spinning black and white thing can take any perceived color, depending on the rotation rate. In their case, the distance between the plates modulate the light color. Once a ray leaves the screen, it is of a given color and won't change anymore.
What I didn't see is the issue of lighting the surface. This needs a front light. Put the technology has one main advantage: it can emits any visible frequency. Hence, its gamut should be much larger.
J.
I don't quite think the poster understood the article. From the article:
Once a voltage has been applied to an iMoD element, it requires less power to hold the metallic layer in place than it does to move it.
Looks to me that *some* power is still required to keep the display going. If it loses power the layers would go back to their default state (which while the article does not state, it would appear its white when its off).
Likewise this statement:
but the cool this is that the display actually works like RAM (it retains its state until voltage is applied to reset it)
I'm no RAM expert but from my understanding (with current RAM), as soon as power is lost, so is the data. Unless you're talking about old magnetic RAM from the 50's and 60's, or IBM's upcoming MRAM, but I seriously doubt you were thinking of those.
//m
but the cool this is that the display actually works like RAM (it retains its state until voltage is applied to reset it)
Cool, some people will get to watch their BSOD's a few seconds more.
On a serious note, I wonder if this could actually cause video card makers to make cards that use memory that does not have to be dynamically refreshed, since the monitor pixels can hold the image. Might reduce memory latency for the frame buffers of the future.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
The power of Iridigm displays derives from the replication of some of Mother Nature's most beautiful creations: Butterflies.
Obliteracy: Words with explosions
There are millions of CRTs out there helping businesses make money. Now these CRTs and to a lesser extent LCDs are also costing companies money through mainly power costs. There are also some health and safety issues that cost money through the running of lighting and cost of fixtures and fittings, but we'll let these out for now.
So, where do you have a CRT monitor and an application environment where high performance in the frame rate isn't an issue? Hmmm, how about every call centre in the world. If an IT manager sees the cost benefits of getting low power consumption monitors he or she will bite. If an accountant sees the numbers they'll bite the arm off the salesman. I can see these taking off in a big way with Call Centres and programming shops.
There's a market there for these things, I'd like to see how they do with CAD/CAM apps too.
sic transit biscuitus
This technology is great for displaying text (and pictures of butterflies) but it is very bad for games.
Look at the description of how it works. The colour is determined by the distance between glass layer and the metal plate. Big gap = red. Small gap = blue.
This is fine for static images, but it means that it takes 5 times as long for a red pixel to change state as it does a blue one.
When you have a quickly moving image, the result in severe ghosting for red objects. White objects will leave a rainbow trail - red at the far end, blue near the object. Blue objects are relatively unaffected.
If you do use this for playing Quake 3, just make sure you're on the blue team.
How does this compare to OLED displays, which are super cool. We've all been waiting for something without the pitfalls of LCD. This looks cool too. 400 - 1000 dpi? SWEET.
Resisting LCDs until OLEDs or this Iridigm thing is like resisting the tape cassette and listening to vinyl until CDs came out.
sig
You have to appreciate post-Dot.Com tech reporting:
provide breif overview of how new technology actually works - consult glossy side of start-up's brochure/PowerPoint presentation
Thank you c|net for providing us all with that fine peice of tech journalism. Too bad Richard Shim couldn't fill more copy space by staring at Maria Bartiromo on CNBC, and had to resort to describing technology halfway through the article.
credo quia absurdum
IANAEES....
Both SRAM and DRAM require constant power to reliably store data.
SRAM differs from DRAM because the cells that hold bits are always charged [howstuffworks has a diagram, basically its 5 logical gates in feedback]. As a result SRAM takes more power but has no refresh delays [and is bigger]
DRAM uses capacitors to store the data and requires refreshing. This makes DRAM smaller, less power instense but much slower.
For example, cache inside processors is a version of SRAM. If SRAM were as cheap as DRAM we'd be seeing 2MB caches common place nowadays...
Anyways... Peace out.
Someday, I'll have a real sig.
Overview and demonstrations of these are available here ->
Universal Display Corporation and Koda Research
Analytic & algebraic topology of locally Euclidean meterization of infinitely differentiable Riemmanian manifold
To solve the problem of undesired residue on the screen the manufacturer could add a slider on the bottom of the unit that the user would slide from one side to the other - erasing the content :).
-- &&
their primary focus is "mobile phones, Smart Phones, Personal Digital Assistants (PDAs), two-way pagers, game players, and other mobile appliances". It could be that these displays are impractical for some reason (perhaps fabrication) in larger sizes. As usual with technology like this, the real issue is scaling production.
If this tech can avoid dead pixels it would get my money. hell, I'd pay a premium for a flat display with no dead pixels. I just go a new computer that came with a LCD monitor and it has a dead pixel. I find it very distracting. I set the colors on my monitor dark to minimize eye strain and a bright white pixel glares at me. I loathe it. I use my CRT when I have serious work to do. Is there anything I can do about to minimize the distraction other than making my monitor look like I'm staring a lightbulb??
Reality is that which refuses to go away when I stop believing in it. --Phillip K. Dick (remove SPAM to email)
Actually, the hysteresis in the MEMS position suggests that a residual image might be maintained if power is lost. It just won't retain the original colors.
How can we afford to ever sleep
So sound again
--ebtg
IIRC isn't this a property of Light Emitting Polymers? At least not the first incarnations, or the later revisions in that a charge is only needed to change the polymer state... so more power is used when viewing a constantly changing images (i.e. multimedia), whereas spreadsheet/office use would be on the lower end of the power scale.
Are you local? There's nothing for you here!
It sounds more like "e-paper" than a LCD replacement. It doesn't produce it's own illumination.
If you actually look through their site, it looks like they are aiming for the PDA market, not the desktop display. Perhaps a limitation of the technology, perhaps a really good understanding of the strengths and weaknesses of their product.
Interesting that the site spouts off on touch screen technology. I've always loved the spontaneous change of LCD to LSD when you press on you LCD pannel, with these, you might just semi-permenantly change the pixel!
And they are showing progress, definitely beyond the "vaporware" that some commentors have said. It appears that they *have* a working product that they demo'ed in May of 2000.
Iridigm Demonstrates First Color iMoD Matrix(TM) Display
SAN FRANCISCO, Calif. - May 20, 2002 - - Iridigm(TM) Display Corporation, a developer of flat panel displays for mobile devices, will demonstrate its iMoD Matrix(TM) technology at the Society for Information Display (SID) International Symposium in Boston, Massachusetts. During the Exhibition portion of the conference held May 21-23, 2002, Iridigm will demonstrate the color iMoD Matrix(TM) display in its booth #1805/1807. This is world's first direct-view color flat panel display based on MEMS (Micro-Electro-Mechanical-Systems).
Continued here
www.christopherlewis.com
It needs a front light but only in dark environments. Apparently, the reflectivity of the surface is sufficient for normal lighted environments
How can we afford to ever sleep
So sound again
--ebtg
but it was good of you to think of the modulation rate based color method. BTW, did you know that modulation based color perception is a genetic trait? not all people percieve color from the spinning disk experiment. i am one that does not, and i was very frustrated when i was trying to get the experiment to work until i found out that some people are not sensitive in that way. folks in my computer club were programming their B&W monitors to show color using the technique before there were any color TV interaces.
The problem is that each time the metal is bent it does still go though some changes on a microscopic level. If you only bend it X% then it will last for a long time, but that doesn't mean it will last forever. You will still get changes in the atomic matrix (migration of the atoms, the atomic structure changing from one form to another, micro-fractures, introduction of foreign materials) at the points of stress, enough so that eventually the metal will break at those points.
Sapere aude!
How do they get color graduations? If the plates are bistable, how do they get more than 8 colors (with each of red, green, and blue being either on or off)? Or can they make the plates hover in-between stable states by applying a current?
er...
dere's around 100 cells per pixel, so you night get significantly more than 8 colours...
oh brave new world, that has such people in it!
If it were like RAM, voltage would have to be applied to maintain the display.. removal of voltage would mean loss of data. Did you mean EEPROM or Flash?
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
Besides RTFA, the described interference pattern is the same principle that holograms are based on. They work.
I heard on NPR the other day an even neater sounding alternative that is about five years off.
_
It uses the fact that certain plastics when charged with electricity will emit light and certain colors. The screen would be flat and completely flexible.
Literally you would have a screen (a TV for example) that could be rolled up and put into your backpack.
Right now they are looking into small scale electronics applications of the technology in terms of putting in screens for car radios and such but they have the big plan of a flexible plastic tv or computer monitor.
Of course if you pay attention is the fact that it needs no backlighting and can be extremely thin. Very neat stuff.
_______________________________________________
ACK
Since this display relys on the interference of light, would it work in the dark? With no light to interact with it wouldn't display. A backlight wouldn't seem to do the trick since the refraction has to be toward the user? I'm not a display guru by any stretch of the imagination, so maybe someone can explain this to me.
THIS SPACE FOR RENT
They claim that since the entire display is inorganic, it's insensitive to temperature variations. Looks like the marketing folks have gone a bit too far on this one. Metal and glass have very different coefficients of thermal expansion. That suggests that the metal layer will be under tension at cold temperatures and under compression at high temperatures. This should affect the interference layer thickness achieved at a particular voltage. I expect that this will, at the very least, affect the display colors since interference wavelength is very sensitive to the thickness of the interference layer.
Anyone care to do the math?
How can we afford to ever sleep
So sound again
--ebtg
(I might take this time to note that screen savers don't really have a place on a modern desktop other than eye-candy. But hey, I like eye-candy too.)
You can't get a blue screen on a black and white monitor.
I'd really like to have some photographers chime in on this one.
...why?
I'm a photographer myself and "amateur" would be an understatement. I've always been vexed by the inability of the camera to record what I see. For example, I went to the Boston Aquarium a few months back and while my shots were acceptable, the colors were nothing like what I was seeing in-person. Brilliant blues and yellows look painfully muted and boring in my results. I'm told that is a shortcoming of the photography medium and photographers have to use tricks to get those wonderful colors you see in mags like National Geographic, Photo, etc. Well
So what I guess I'm asking is "can this technology be used to not only create and present colors in a 'natural' way but possibly capture them that way as well?"
My
Limekiller
While this is likely true (I only did introductory materials engineering, but it makes sense), some (many? all?) metals have the interesting property that if you leave them long enough after bending, they will `heal' and you'll be able to bend them again. Of course, different metals have differing heal times, with lead and gold being quite fast (hours to days?) and most others being quite long (in the months to years range?).
Bill - aka taniwha
--
Leave others their otherness. -- Aratak
SOunds like interesting tech. What I find interesting is that they have to join several of there elements together to make a "pixel" - presumably this is to avoid having to scrap every video driver in the known universe. BUT it also means they could run at much higher definition if each element is controled seperatley. Combien that with the more paper like look and you have somthing much much easier to read.
Question though - I may have missed htis but how efficient is the manufacturing process? Isnt the main problem with LCDs that the manufacturing process is incredably inefficient?
<fnord>OBEY</fnord>
Looks like they might be giving up some of the lower voltage benefits in order to get higher pixel density. Hence their claims about glossy magazine appearance?
How can we afford to ever sleep
So sound again
--ebtg
You bring up an interesting point: it's not clear how a device like this can produce different saturation levels for a pure hue. In other systems, a single subpixel has a single color but variable intensity, and subpixels of different colors can be combined to produce a range of colors. In this system, each subpixel is capable of producing any color, but only at an intensity defined by ambient light. Consider a three-subpixel unit where each subpixel can be either white, red, or black. This gives only the following possibilities: white, black, two shades of grey (BBW, BWW), and six kinds of red (RRR, RRB, RRW, RBB, RBW, RWW). Now, a single subpixel could be cyan or indigo all by itself, creating a different kind of flexibility, but I'm not sure if that's as useful as what we get with variable-intensity RGB subpixels.
Slashdot - News for Herds. Stuff that Splatters.
will still crack and break for no apparent reason.
On the website, it's stated that since the manufacturing process is done at a low temperature, plastics can be used in a future implementation of the product. This would mean that the display would be more durable than LCD displays with glass. Currently they use glass in the manufacturing procees, probably due to the tight manufacturing integration with LCD display manufacturing, since they're trying to reduce initial costs.
If you're looking here for something insightful or thought provoking, you're probably looking in the wrong place.
What I said was not so much an analogy as an invitation to use your imagination. Admittedly, it is much harder to make a roll-up screen than a keyboard. However, a monitor that shared many of the properties of paper would obviously be useful. Flexible monitors are not ludicrous like a drivable blender and your other mockeries.
In fact, I'll go out on a limb a little and say that such technology will be invented during my lifetime. I have a while yet to live after all.
Tell us why it's better then OLEDs.
Everyone knows LCDs are just a stop-gap, 10 years from now everyone is going to be using OLEDs.
I suppose the benefit of these things is that they are reflective, and thus probably don't require much power when they don't change... But the structure looks pretty complex compared to OLEDs.
If these guys have a very short timeframe to production, they might make out pretty well, but 5-10 years from now this tech won't really be all that relevant, IMO.
autopr0n is like, down and stuff.
SRAMs can be designed for raw speed (CPU caches) or low power (CMOS memory in old PCs before flash). High speed SRAMs can suck down a lot of power due to all of the gates and frequent logic transitions.
OTOH, The low power SRAMs intended for nonvolatile storage use all CMOS FET transistors in their logic gates. These gates draw essentially zero current unless they are actually switching.
Thus, while low power SRAMs require a voltage (typically supplied by a battery) to retain their state, they draw no current when idle. Therefore, in a technical sense, they don't actually require "power" (voltage*current) to keep their state, just a static potential.
A hydraulic analogy would be rigging two toilet flush flap valves in series, then ensuring that they never open simultaneously. This setup could store one bit (1 - open/closed, 0 - closed/open) with just static water pressure and zero flow. (A little water would flow when the valves are actually flipped.)
(btw, IAAEE)
Yes, this does present a serious problem to the technology.
Although they may have many of these in each 'pixel' as mentioned earlier, the circuitry required to drive these sub elements to give each pixel even a modest pallette depth would be absurd (IMHO).
For good (24 bit) color (8 bits per gun), you'd need 768 (256x3) sub pixels driven with at least 768 times the number of connectors to the display and 768 times the bandwidth, or you'd need to have integrated decoder/driving circuitry for each element. You couldn't just send an analog signal as with an LCD. There IS NO ANALOG DRIVING in this device, period.
Now, if the the 'memory' of these devices is truely bistable, they may be able to achieve usable bandwidth using a good multiplexing, but the size of the sub elements along with the rediculous number of connectors per pixel is a SERIOUS issue and the fact that there is NO MENTION of driving levels gives me serious reservations.
IIRC Static ram is faster then then regular Dram, but requires a lot more hardware per bit. Static ram also retains its state when the power goes out. Maybe the framebuffer on the display uses SRAM?
autopr0n is like, down and stuff.
Hook it up to a cellular network and they can download new ads into it....or even better, the states could have an emergency warning/traffic system to take over the billboards when needed...endless possibilities.
//m
Since the color is controlled by varying the distance between the plates how is gamma controlled? How do you make a dark red and light red? Or grayscale. The article doesn't seem to explain this.
Uh, no. Framerate in general has nothing to do with the actual display, although the image will look better if the two are in synch at some multiple.
Framerate, at least when you're talking about gaming, is how fast the game engine and graphics card can update memory. The refresh rate is how fast the electron beam is swept across a CRT. LCDs don't have refresh rates, but they do have response times And I would assume this thing would as well.
The "frame rate" on an LCD or one of these things is 1/response time.
autopr0n is like, down and stuff.
Well, the Economist article reproduced on the Iridigm website indirectly answers this question.
Iridigm's technology, which it calls an interferometric modulator, or I-mod, works by fine-tuning the gap between reflective surfaces. I-mod pixels (the dots that make up the display) are tiny paired mirrors, and the distance between these mirrors can be adjusted to one of four settings. Three of these settings correspond to the primary colours red, green and blue, from which all other colours can be constructed. The fourth is "closed", which means that no light can be reflected, and so the pixel, and thus that part of the image, is black.
Apparently, they'll be generating all colors (including white) by using RGB (plus black) combinations
How can we afford to ever sleep
So sound again
--ebtg
If SRAM were as cheap as DRAM we'd be using it for system memory and might not even need cache at all.
Not necessarily. There's an inherent slow-down associated with large address spaces. Not to mention the heat decipation. Heck, why else do we have 3 to 5 layers caching? The practical approach is to have successive layers of cheaper, larger and slower memory.
Since we already have 8 meg caches (in some high end machines), there's little value in doing away with multi-gig low-power, low-cost memories. Theoretically some apps will achieve noticable performance gains, but at enormous costs (today at least).
Furthermore, DRAM with internally managed refresh logic is functionally identical to SRAM (but non-deterministically slower). For something like video memory which regularly touches every byte of memory, the refresh logic would be unnecessary; thereby speeding up the memory. Further, DRAM is sufficiently performant enough to handle refreshes. 4MB * 80fps (for true color 1280x1024) = 320MBps. DDR can handle 2.1GBps alone. This doesn't even acknowledge the possibility of interleaving/banking/segmentation or what-ever types of tricks they may utilize.
-Michael
the eye strain isn't caused by interference with 60hz the power circuits. Modern monitors have a lot of protection from things like that.
The problem is that you can see the image blinking on and off, and it's annoying. I can still see flicker at 70hz, and in general prefer something in the 80s.
autopr0n is like, down and stuff.
This could be a real display revolution, and no I don't work for them. They are talking about print quality displays here. They don't mention frame rate or gamma, but my guess is that gamma is a function of how many subpixels remain black, and if the frame rate does turn out to be low give them some time. This is v0.
Maybe you'll need to iron your screen every so often?
Well, an iron probably wouldn't get hot enough, and would likely scratch the screen. But some adaptation of that might work.(Probably built in, like a degauser.)
I think we've pushed this "anyone can grow up to be president" thing too far.
(isn't TV interlaced 50 FPS?)
NTSC TV, what we use here in the US is 60 feilds per second (a feild is half a frame, every other line)
HDTV has i and p modes, i modes are the same as NTSC as far as speed, and p modes redraw the same frame each time... IIRC
autopr0n is like, down and stuff.
You can have a diffrence in voltage without constant power. I don't know if it would work for this type of thing.
autopr0n is like, down and stuff.
I Am An Electrical Engineer?
Blah Blah Blah.
So what I guess I'm asking is "can this technology be used to not only create and present colors in a 'natural' way but possibly capture them that way as well?"
This question is like asking if a new kind break technology for your car will make your car edible.
autopr0n is like, down and stuff.
Most metals exist in more than one form of crystal matrix. These different types of crystals exist in almost every chunk of metal you find. You will usually end up with a small area of one form of crystal (with all atoms lined up in the same direction) which is surrounded by another form of crystal. These small areas are called grains. The smaller these grains are, the more easily the metal bends, due to the fact that the atoms on the edge of a grain do not bond well to the atoms outside the grain.
When you bend metal you tend to form more grains in it, due to the movement breaking up existing grains and splitting them into smaller pieces. The increase in grains causes the metal to weaken, even if it is a small amount every time. If the metal is allowed to "relax" for a period of time, there is the chance that two extremely close and aligned grains will convert the atoms between them into their crystaline form. This reduces the amount of grains and re-stiffens the material. This re-conversion is very slow under normal temperatures and pressures and thus is a minor effect.
You can increase the grain size and lower the number of grains by heating the metal at a certain temperature for a period of time. If you then quickly cool the metal (quench it in water, for example) you will end up with a harder material (but more brittle). This is how blades are made that hold an edge and stay sharp, the harder the blade is the better it will hold an edge. However, if you make the blade too hard then it will not bend at all and it will be brittle.
Sapere aude!
make these out of 'real' butterflies?
Creationists are a lot like zombies. Slow, but powerful and numerous. And they all want to eat our brains.
I guess I read that wrong. In that case, he's full of shit. For one thing, incandescent lights don't flicker, and for another, two signals at the same frequency won't cause other patterns.
It also wouldn't explain why monitors would flicker with the lights off, either.
autopr0n is like, down and stuff.
My understanding was that the "Static" in SRAM meant static ciruit design, meaning no current usage in the idle state, meaning static CMOS gates as you mention. But you say that this only holds true for the "Low Power SRAMS"
So how are the circuits set up in these "high speed SRAMS" so that they draw idle current?
I was under the impression that *all* SRAMS use static cmos circuitry.
The following sentence is true. The preceding sentence was false.
"If SRAM were as cheap as DRAM we'd be seeing 2MB caches common place nowadays..."
:)
If SRAM were as cheap as DRAM I'd hope to see machines using it as system ram
While the method of color control was clearly explained, the article didn't explain where the actual source of light would be. Will it be behind the display? Will it be shone onto the display? You can't generate light by simply providing a resonating chamber, which is what they seem to be doing.
Light generation seems to me to be the biggest draw of current.
What's this Submit thingy do?
Well if you had looked at the technology at their web site, the subpixels CAN'T be White. Thay can be only Black or one specific color (eg. Red). The technology has the color of each subpixel fixed by it's physical properties. (ie. it's digital, not analog) so there are red/black, green/black, blue/black subpixel types. Same idea as the old CGA displays but without the intensity bit (ie. only one bit per gun.)
Sadly, no mention on the site of how they think they play to provide any INTENSITY information.
VERY funny. Yes, I understand how RGB monitors work. However, with optical interference you can get a fairly continuous selection of wavelength (i.e. color...at least the dominant aspect of color) as long as you have fine enough control over the layer thickness.
How can we afford to ever sleep
So sound again
--ebtg
ECL was fast, but it was just about as opposite of CMOS as you can get. It works using bipolar transistors to continually shunt large currents through resistors even when the gate is idle. That single 1K chip I worked on probably drew several of watts of power. Nevertheless, it was considered to be a SRAM.
(The mainframe CPUs put a hundred or more ECL chips on a ceramic substrate, then used the mother of all water cooled heatsinks to pull out the massive heat that was generated.)
If iHear another name with some iPrefix, iThink iMGonna puke.
Accountability on the heads of the powerful.
Power in the hands of the accountable.
As I understand it, the reason for sub-pixels isn't to avoid blowing the minds of video drivers, but to create better colors.
One difference between this and other display approaches, as other posters have pointed out, is that each 'element' must be set to a particular color AND intensity *during manufacture*.
In a CRT, you only have to choose the color, and can vary the intensity on the fly.
So, you need a group of pixels set at different colors in order to create the 'light purple' vs 'dark purple'.
This wasn't clear in the article, but I think its correct, based on what other posters are saying.
-Zipwow
I don't know which is more depressing, that 2/3 didn't care enough to vote, or that 1/2 of those that did are crazy.
Even an all-optical solution to memory storage wouldn't eliminate the advantages of(or, for that matter, the need for) having on-CPU cache.
We use on-die cache because CPUs operate at a datarate high enough to justify keeping frequently-used data close at hand. It's justified for the same reason that you'd want to use one Athlon 2800+ CPU, instead of a multiprocessor system using the same level CPUs, but at half the clock speed.
What's this Submit thingy do?
That makes it even more limiting, doesn't it?
If you had read the web site, you'd know that the color of each pixel is determined by the size of the gap (that's why blue has a smaller gap than red) and the resulting interference. I'm also making an optimistic assumption that they'll figure out a way to make the gaps variable rather than strict on/off.
...which was exactly my point. Thanks for playing.
Slashdot - News for Herds. Stuff that Splatters.
Ah yes, you finally grasp the point of my original question about how they were planning to generate white light from a single pixel when they had access to only single colors via interference (white light being generated by the simultaneous presence of multiple frequencies/colors of light)!
Since then, I've realized (as you've pointed out in other threads) that they're using only bistable pixels and not taking advantage of the possibility of continuously variable interference layer thickness, so my original question is moot, anyway. :(
How can we afford to ever sleep
So sound again
--ebtg
Anybody else getting that stupid iPAQ pocket PC ad over the top of CONTENT on the page? I'm using Moz 1.1 in OSX 10.2
Yes, it's a blog. Sorry if that offends you.
You'll have to excuse me, I was shooting from the hip and didn't realize that I had made a mistake in my original discussion.
I originally said, "When you bend metal you tend to form more grains in it, due to the movement breaking up existing grains and splitting them into smaller pieces. The increase in grains causes the metal to weaken, even if it is a small amount every time."
This is not exactly true, it had been a while since I studied metallurgy and I didn't have any reference texts to consult. To clarify, the reason the metal weakens is not that the number of grains is increasing and making the material more ductile (easily bendable), but that the dislocations (areas of stress in the metal matrix) and impurities are getting moved to the edge of the grains and are collecting together. This means that less of the metal has flaws distorting its structure and is therefore harder. Since it is harder it is now less flexible and more brittle. This causes micro cracks to form during the bending. Eventually these cracks lengthen and the metal fails.
Work hardening occurs when the metal is plasticly deformed. These deformations cause impurities and other strains to gather together and less distort the structure of the metal. Since more of the metal is ordered, it is harder than it was originally.
One thing you should know is that metallurgy is very complex. There are many factors which enter into the equation, such as grain size, alloys, impurities, many different phases (crystal structures) of the metal, etc. Often simply how the metal is composed, heated, cooled, worked can vastly change its properties.
Here are some sites to study more about metallurgy:
PLANT MATERIAL PROBLEMS - a site on metal failure
Metallurgical Terms Made Simple - a site on the basics of steel metallurgy
The Metallurgy Of Carbon Steel - a more in-depth analysis of steel metallurgy
Sapere aude!
So would you rather they called it an "Interferometric Modulator Display"? They've got a good reason to call it iMoD. It's what we call an "abbreviation". Quit yer whinin'.
Why Your Next Computer Display Might Be an Empty Box
By Robert X. Cringely
All the creatures will die, And all the things will be broken. That's the law of samurai. (Jubai, 1605)
Yeah, that's great when you can control brightness.
Have you done the math on this? If you want 24-bit color you'll need 768 subpixels per pixel (256 for each of red, green and blue). That drives up the manufacturing cost and drives down yield (as allowable stuck-pixel counts are exceeded, and there will be stuck pixels just as there are in LCDs) but that doesn't even matter. Those 768 subpixels will form a square 28 pixels on a side; at the stated (minimum) size of 25 microns that's a pixel 0.7mm across - a.k.a. 36dpi. Besides being a generally crappy resolution, that's coarse enough that the color variation within the pixel will be visible to the human eye. Dropping down to 16-bit color gets the pixel size to reasonable (though still not particularly good) levels, so it's possible to have a display that's fine for regular use but will still be shunned by many users. There's also no mention of color-change latency, which might also be a concern for many users.
I don't mean to be a nay-sayer. Personally I'd be happy with 12-bit color and a pretty high color-change latency. My point is really that taking full advantage of this technology will require a fundamentally different approach to color and not just a naive "throw more subpixels at the problem" approach.
Slashdot - News for Herds. Stuff that Splatters.
We shalt bow down to the almighty Google (unless we're a stupid company with a suit against it)...
The Google search, and some of the Google results Like this, and this one.
Although I'm still trying to find the actual strip image... DOH!, I underestimated the almighty Google and it's 'images' search!! Here is the image.
Are you local? There's nothing for you here!
Do you really think holograms work? I got a little one that was being handed out in front of a strip club and let me tell you these thing lot absolutely nothing like the real thing.
Won't this result in the same problems the Game Boy Advanced has?
One of the major things I like about my laptop is I can use it at night...
What's this Submit thingy do?
Since this technology works on light interferance, it appears that it needs a light source. Something not mentioned on their web site.
It's pretty hard to see a butterfly in the dark, I'll bet these displays will have interesting color issues when the ambient light changes.
- Zav - Imagine a Beowulf cluster of insensitive clods...
Sometime in the near future, we'll be able to get a gig of DRAM-based memory for under $40. Even if SRAM takes *eight* times as much silicon real estate to produce, that's still 640MB of SRAM for under the standard-cost-of-computer-upgrade of $200, and with in mind that 640MB "should be enough for anyone".
Please consider making an automatic monthly recurring donation to the EFF
"...so what do you see when the driver crashes?"
Hmm. Which operating system do YOU use?
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
Actually, although it may sound like BS, it seems like the R&D team is the one that may have ingested bodacious plants. Whatever works to spark creativity, technical or otherwise.
Hmmm, How will it work in the dark being based on light reflection? Does it need a minimum amount of light?
Frankly, it all depends.
Transistor-based SRAM retains its state, capacitor-based DRAM loses its state unless you refresh it constantly.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
With this technology, I can get any color I want by varying the plate distance continuously! I can also get (visible) black by varying the plate distance to one extreme or the other. So the red-green-blue thing is a trichromatic herring: to get a given intensity of a given hue, I can just dither between pixels of that hue and black pixels. This doesn't require many bits. For pastels, I need four pixels to dither with, but they don't have to be rgb: I can play with colors that combine in the right proportion against the human eye response curve.
Better yet, humans don't distinguish colors that are very light or very dark pastel very well, so a lot of my dither space can be effectively fudged.
The upshot of all this? I can get a lot higher effective bpp by dithering with this thing than I can from dithering with a fixed RGB color palette, and that means that I can use fewer bits of dither per pixel to get a wide range of colors. In principle.
What's wrong with "IMD"? It's a proper acronym, and doesn't look like 90's ELiTE h4xx0r scr1p7 in print:
iT hAs SaNe CaPiTaLiZaTiOn.
And it fits right in these with LCD, CRT, VGA, CPU, AGP, PCI, ISA, AXP, USB, and all those other fun unpronouncible 3-letter acronyms.
If one must abbreviate instead of acronymalize, at least "IMoD" would presents consistant case.
Kid-proof tablet..
Light reflects off two surfaces, one just beneath the other. If the distance between the surfaces is such that the reflected light waves are perfectly out of phase, the waves will cancel eachother out, making it look like the surface actually absorbs that frequency range, producing color. That means that the distance the light travels between the plates is absolutely crucial in producing the right color. That's why butterfly wings shimmer. Your eyes are each viewing the wing at a different angle, each seeing a different color.
When light hits the plates striaght on, the light travels a certain distace between the plates. But when light hits at an angle, it travels slightly farther, depending on the angle. So, for example, instead of being out of phase at 600nm, light at 620nm will be out of phase, making a different color appear if you look at a different angle.
So unless I missed something, what we'll end up with is a display that "shimmers" like a butterfly wing. The hue of the display will shift when the screen is angled. That means that the effective viewable angle will suck a lot more than it does for LCDs, and it will be almost impossible to be perfectly sure what color you're looking at (particularly important for desktop publishing).
Perhaps someone who knows more about physics can explain how they intend to make this actually work. For now, though, I'm going to wait till I see a working prototype before I sell the farm to invest in their product.
"With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea...."
RFC 1925
What, were you asleep when we went over that? A single subpixel can only be a pure hue or black. Right now that subpixel is limited to a single hue, on or off. Maybe - maybe - some day a single subpixel can vary its hue by varying the gap, but there'll still be no way to vary its intensity. You have one axis of variation (the gap); varying both hue and intensity would require two axes.
Do the math. I just did, and another poster did, and we both came to the exact same conclusion, and yes it does require a lot of bits if you're dithering between pure hues and black. If you could vary a subpixel's hue and you were willing to display only darker shades but not lighter, you could make do with a lot fewer pixels, but that's simply not going to be satisfactory for general use.
Slashdot - News for Herds. Stuff that Splatters.
You misunderstood considerably my 'scrolltop' idea, I think because it is hard to describe. The screen is supposed to have a rigid piece on the end (think so you can pull the screen out of the scroll). Once it was pulled out - this is an important part - rigid braces would fold out of the base of the scroll and attach to the rigid piece at the top of the screen, thus forming a square to support the screen.
The base of the scroll is supposed to have adjustable feet to prevent the thing from rolling away when closed and to stabilize it when the screen is out and up. Really, I intended for people to use this device is a screen-keyboard configuration that is very similar to current laptops. Therefore the screen and keyboard would not come out at opposite ends of the scroll, but instead at closer to a 90 degree angle.
See what I mean now? It is just like a laptop, except it rolls into a scroll, and has some extra uses, like writing on the screen with a digital pen, retaining images on the screen after it turned off, etc.
Size would make or break this thing - if it were significantly more compact than modern laptops when rolled up it might be nice. But this is just a silly idea.
on a tangent though, if the iMoD display retains state with the power off, why not use it as nonvolatile memory?
To a human it looks like it's displaying snow... but to a high res camera (or some hardware that can read the pixel's state without resetting them) that's actually some of the the data in your hard drive.
I remember that flash harddrives are said to only have a million or so read write cycles, even with wear leveling. since the marketing flacks say that the iMoD can survive years of rapid pixel flicking, using a derivative of its technology for semi-optical or electrostatic harddrives might be a better option.
Not very portable compared to the book-sized SIMM-eating wonders of today, but potentially very bright and certainly a damn sight cheaper. Advertisers would love it; now you can afford to have your logo take up the entire side of your building, at least at night.
Got time? Spend some of it coding or testing
Based on the -- admittedly anecdotal -- evidence of several hundred of the people I game with in Aces High, a MMOACS (Massively Multiplayer Online Air Combat Simulation) put out by HiTech Creations, up to the point where your frame rate reaches the refresh rate of your monitor, increasing the frame rate directly contributes to your success; the smoother and faster your display refreshes, the better you are able to follow the maneuvers of the aircraft you are attacking. I know that I can tell the difference between the 57fps I get when flying around without anyone else visible and the ~40fps I get when I'm in a furball at low altitude over an airfield, with eight or nine other planes, ground clutter, structures, and the tracer trails from gunfire.
The Holy Grail of game development is to keep the frame rate high enough that it stays above the eye's fusion rate at all times. Unfortunately, because of the wide variation in the hardware the game will run on (for PC games), this is functionally impossible to achieve. That's where console games have an advantage -- because the game designers know what hardware the game will be running on, they can optimize the hell out of the code for that hardware. That's why most console games look better than the same game on a PC -- the display code isn't as heavily optimized, so you need more power to get the same frame rate.
The only real advantage that the iMoD display is going to have over a standard CRT is being able to get a displayed frame rate that matches the generated frame rate without having to worry about vertical sync to prevent artifacting the display. Unless it's a quantum leap in display speed over CRTs, and doesn't suffer the 'dead pixel' manufacturing problem that keeps big LCD displays commercially unviable (yield problems, etc.), it's likely to wind up not getting the sales volume it needs to pull it up out of a niche market. On the other hand, if it is faster than CRTs, it's going to have a solid niche market to the high-end gaming community.
Keep in mind that even a 1600x1200x24bit display is only holding 5.5MB raw.
While the physical media remains in one state under a low-power mode, I highly doubt that much effort has been put towards reading the state of the physical media.
Interesting, though, that this technology could very easily be scaled to a much deeper bpp range. Say 16 bits per color, and you're now sporting 11MB raw.
But face it; do you really want someone to be able to walk up and photograph your 5.5MB of your data?
Though you might be able to store some data in the least significant bits, though. Say you have 16 bits per color; You can get 8 bits per color quality even if you dedicate the lower 8 bits to data storage.
I'd be interested to see some links on the flash data, though.
What's this Submit thingy do?